The Birth and Evolution of the Concept of Panayiotopoulos Syndrome

“I can see nothing,” said I, handing it back to my friend. “On the contrary, Watson, you can see everything. You fail, however, to reason from what you see. You are too timid in drawing your inferences.” The Adventures of Sherlock Holmes: The Blue Carbuncle This issue of Epilepsia portrays eminent contributions (Caraballo et al., 2007; Ferrie et al., 2007; Koutromanidis, 2007; Martinovic, 2007) on a common childhood seizure susceptibility syndrome that primarily involves the autonomic nervous system (ANS). I have been honored with its eponymous nomenclature and the editors’ invitation to write a personal note with the above title. It all started in 1975 with two otherwise normal children, A and B, who had EEG occipital paroxysms (Fig. 1) but different clinical manifestations (Panayiotopoulos, 1980, 1981). I saw them in my part-time private practice and followed their progress for over two decades (Table 1). EEG samples from the original studies in Athens. Top: Occipital paroxysms of child A and B. These are runs of continuous spike and wave complexes in the posterior regions. In routine EEG they occur as long as eyes are closed and are eliminated when eyes are opened. It was documented that the activation of the occipital paroxysms was due to the elimination of central vision and fixation (left of the vertical bar, symbol of eyes with glasses) and inhibition by fixation (right of the vertical bar, symbol of eyes without glasses). Bottom: EEG of children with PS but without occipital spikes. My scientific curiosity of why occipital paroxysms occurred only when the eyes were closed was satisfied with a series of simple tests that documented the activating role of the elimination of central vision and fixation, which I called “fixation-off sensitivity”(Panayiotopoulos, 1980, 1981, 2002). The clinician's responsibility of making an accurate diagnosis was more difficult to satisfy. Child A had brief visual occipital seizures or blindness often followed by headache (Panayiotopoulos, 1980; case 24 in Panayiotopoulos, 1999a), which should be differentiated from migraine. Child B was a different diagnostic problem (case 1 in (Panayiotopoulos, 2002). In early childhood she had two nocturnal episodes of vomiting and eye deviation before becoming flaccid and unresponsive for 15 minutes. Between the ages of 7 and 9 years she experienced infrequent, brief diurnal attacks of dizziness and visual illusions sometimes followed by impairment of consciousness and automatisms. Experts that I contacted were as puzzled as I was and occipital epilepsy appeared to be a neglected topic in the literature. Therefore, I initiated a prospective study of occipital epilepsy with gratifying results mainly regarding the differentiation of occipital seizures from migraine and the formulation of Panayiotopoulos syndrome (PS). A consistently emerging pattern was of a characteristic clustering of otherwise normal children with infrequent, mainly lengthy and nocturnal, seizures associated with vomiting, deviation of the eyes, and excellent prognosis. Eight children strictly fulfilled all these criteria, which I thought constituted a new epileptic syndrome, and I submitted a report for publication in 1985 (Panayiotopoulos, 1989a). However, this new syndrome was much wider than my initial evaluation and the supposition that it was an occipital epilepsy because of EEG occipital spikes proved largely incorrect. Following a comprehensive appraisal of 18 children with occipital spikes, the concept of “early onset” (16 patients with symptoms similar to child B) and “late onset” (two patients, including child A) benign childhood epilepsy with occipital paroxysms (BCEOP) was introduced (Panayiotopoulos, 1989b). Ictal vomiting was a mystery with only a few reported cases mainly in adults with nonoccipital epilepsy. In a retrospective file review of all my 900 patients (Table 1), ictal vomiting occurred in 24 patients, all children, whose symptomatology was similar to the early onset childhood seizures of case B (Panayiotopoulos, 1988). The majority (21) were developmentally normal and from the EEG point of view could be divided into two subsets. The larger consisted of 12 children with occipital spikes, and was classified as early onset BCEOP. The other subset comprising nine children had extra-occipital spikes or normal EEG: With such EEG heterogeneity I did not then consider them classifiable in a uniform syndrome. It was 10 years later that it became apparent that these two subsets constituted the same syndrome because of identical clinical symptomatology and evolution despite different EEG features (Panayiotopoulos, 1999b). This was not a solitary effort. PS exists because of independent prospective and retrospective long-term studies of distinguished colleagues (Caraballo et al., 2000; Ferrie et al., 1997; Kivity et al., 2000; Lada et al., 2003; Oguni et al., 1999; Ohtsu et al., 2003; Sanders et al., 2004), editorials (Ferrie and Grünewald, 2001; Koutroumanidis, 2002), reviews (Covanis, 2006), and consensus statements (Ferrie et al., 2006; Ferrie et al., 2007). Their knowledge, views, and doubts have been fundamental in the evolution of the concept of this syndrome through many years of lengthy personal communications and debates that continue to date. A recent in-depth review of all available data in my monograph (Panayiotopoulos, 2002) reinforced my view of benign childhood seizure susceptibility (a concept unifying rolandic epilepsy, PS, and other phenotypes of benign childhood seizures) and also revealed important previously unrecognized features. “Loss of muscle tone with unresponsiveness” (ictal syncope), a common ictal manifestation (see case B) that may occur alone without other ictal symptoms, has significant clinical implications. Ictal behavioral manifestations, other than emesis autonomic symptoms (ictal syncope, thermoregulatory, vasomotor, or cardiorespiratory) sometimes constitute the most obvious clinical features. PS with seizure-generated fever may mimic encephalitis or febrile seizures; so far secondary hippocampal damage as described following true febrile convulsions has not been reported. In distinction to other focal epilepsies, the clinical autonomic symptoms of PS at onset and the interictal EEG foci are not suggestive of a particular anatomical location. Recurrent seizures may not be stereotyped. Autonomic status epilepticus is common, but brief seizures or those consisting solely of inconspicuous autonomic irregularities are unlikely to be observed and reported. The ILAE task force has recognized PS with emphasis on “early onset benign childhood occipital epilepsy” (Engel, 2006). Where do we go from now? PS with its multiple clinical and EEG peculiarities that long eluded recognition, challenges traditional views and opens new avenues in childhood epileptology. We may have seen only the tip of the iceberg and it is likely that significant pieces are still missing from the jigsaw puzzle. There are many practical, diagnostic, genetic, etiological, electrophysiological, pathophysiological, theoretical, epidemiological, social, and management issues still to be resolved. This will require a multidisciplinary approach. Objective documentation of seizure manifestations is necessary: carers should easily be able to recognize the onset of recurrent seizures, assess basic ANS functions and record the events (many are lengthy) on digital recorders and mobile phones. Researchers on ANS function and ANS nonepileptic childhood diseases should involve this model of an ANS seizure disorder. Modern genetic investigations should be applied to what is probably a genetically determined disease and which may have genotypic links with other clinical phenotypes (Grosso et al., 2007). Epidemiologists should reassess how recognition of this common syndrome might affect the incidence and prevalence of childhood epilepsy in general. Clinical psychologists and other healthcare providers should address parents’ severe and sometimes long-lasting trauma and anxieties (Valeta, 2005). Improving diagnostic yield is a prerequisite for any step forwards and an area that audits would expose. Clinically, any child with ictal emesis or other paroxysmal autonomic manifestations, particularly if prolonged, should be suspected of having PS. Electroencephalographically, any EEG with frequent, multifocal (or both) epileptogenic foci from a normal child with infrequent seizures should raise the possibility of PS. Children with a diagnosis of atypical febrile seizures (particularly if these are not convulsive) or a quick recovery after a hospital admission for possible encephalitis should raise the suspicion of having PS. Reliance on skilled EEG technologists is often rewarding (Sanders, 2004). My enduring plea to the relevant official bodies is to re-examine the evidence and initiate research, scientific publications, practice parameter guidelines and public awareness campaigns as those that eliminated the myths and the anxiety regarding febrile seizures. The situation dictates that action on PS and particularly management should be prompt. Recommendations for prophylactic and emergency treatments are mainly anecdotal, new antiepileptic drugs have not been tested and possible adverse effect of “rescue” benzodiazepines in autonomic status epilepticus have not been assessed. Prognosis is commonly excellent but a minority of children with typical PS, now surfacing with increasing incidence, it seems that they are at risk of frequent and lengthy autonomic seizures that may be life threatening, either because of excessive iatrogenic interventions or seizure-triggered cardio-respiratory arrest. We cannot allow room for delay or error here.